Opinions - Is Intel scared of Nehalem?

[HEXUS]

The oxymoronic-sounding headline may not be quite as crazy as it reads.

Read more.

[j.o.s.h.1408]

right now, intel are competing within themself atm and you maybe right, that initial benchmarks dont tell teh true story on how powerful nehalem realy is. i mean showing off full benchmark numbers on a produc thats coming out in a few months time could do more harm then good with your current stock. i expect proper numbers to appear with weeks of nehalem finaly coming out.

i cant wait for these baby's to come out

[strengthofmind]

The 6th and 7th paragraphs don't quite match up.

Nehalem's performance not being quite up to scratch, but good enough to crush sales of present-generation parts?

[Tetras]

This article:
http://www.anandtech.com/weblog/showpost.aspx?i=480

Has a different opinion, in that it looks like the architecture has been designed with the intention of rounding it out and desktop performance is not expected to leap miles ahead of Penryn until quads are better supported.

[DR]

This article:
http://www.anandtech.com/weblog/showpost.aspx?i=480

Has a different opinion, in that it looks like the architecture has been designed with the intention of rounding it out and desktop performance is not expected to leap miles ahead of Penryn until quads are better supported.

I'm not sure I understand what you are saying - the performance is leaps and bounds ahead of C2D/Q at a clock for clock level...

http://www.hexus.net/content/item.php?item=15015

we have shown performance on the product here....

Could the lack of numbers have something to do with Nehalem's performance not being quite up to scratch? That seems highly unlikely, especially if our Nehalem performance preview is accurate, and we have no reason to doubt that our 2D numbers will stack up against retail samples.

I think if you read the comments - http://forums.hexus.net/showthread.php?t=145760

You can see the general thoughts of the HEXUS reader, some will jump others won't

It will be interesting to see the performance at the lower clock speeds and see how they stand up...

[Tetras]

I'm not sure I understand what you are saying - the performance is leaps and bounds ahead of C2D/Q at a clock for clock level...

I was a bit ambiguous, by desktop performance what I meant was games.

Nehalem is about improving HPC, Database, and virtualization performance, and much less about gaming performance. Maybe this will change once games get some heavy physics threads, but not right away.

Why? Most Games are about fast caches and super integer performance. After all, most of the Floating point action is already happening on the GPU. The Core 2 CPUs were a huge step forward in integer performance (not the least because of memory disambiguation) compared to the CPUs of that time (P4 and K8). Nehalem is only a small step forward in integer performance, and the gains due to slightly increased integer performance are mostly negated by the new cache system.

The percentage of L2 caches misses for most games running on a Penryn CPU is extremely low. Now that is going to change. The integrated memory controller of Nehalem will help some, but the fact remains that the L3 is slow and the L2 is small. However, that doesn't mean Intel made a bad choice. Intel made a superbly good choice by improving the performance where Core (Merom/Penryn) was mediocre to good. Penryn was already a magnificent gaming CPU, but it could not beat the AMD competition in HPC benchmarks, and AMD put up a good fight in database performance benchmarks. Now Intel is ready to fix these shortcomings.

I take that to mean clock for clock up against high clocked Penryns where quads aren't yet supported Nehalem doesn't perform well right now (against Penryn) and this also seems to be the case with your gaming benchmarks (with the exception of 3D Mark Vantage, which can deal with the cores).

[cordas]

The 6th and 7th paragraphs don't quite match up.

Nehalem's performance not being quite up to scratch, but good enough to crush sales of present-generation parts?

I think you have misread the 6th paragraph...

Personally I think there is a fair bit of water in the bucket.... Mind you I think Intel maybe jumping the gun a little bit, yeah people in the know are going to wait for the i7, but we always where.... Joe Public probably isn't that intrested.... computers always get faster, thats a fact of Moore's law.

When you have the cash / need you buy, putting off for the next great part is really a false economy as there is always one coming around the corner thats going to be so much faster.

[shaithis]

Intel have never held off on benchmarks and hype in previous CPU releases. Something smells a little fishy to me......

[j.o.s.h.1408]

Intel have never held off on benchmarks and hype in previous CPU releases. Something smells a little fishy to me......

do explain further what you mean please

[baius]

I was a bit ambiguous, by desktop performance what I meant was games.

I haven't seen any Nehalem benchmarks for Games - and I've been looking. Have I missed something?

[Tetras]

I haven't seen any Nehalem benchmarks for Games - and I've been looking. Have I missed something?

When I mentioned the hexus article I meant this page:
http://www.hexus.net/content/item.php?item=15015&page=8

I just believe the reason is not driver optimisation but the architecture, based on the musings from anand.

[strengthofmind]

Regarding my earlier comment, I think the 6th paragraph has been edited to give the reverse and correct meaning compared to the original. I think the word unlikely has replaced likely. All makes sense now. I should have realised it was a typo.

[cordas]

Intel have never held off on benchmarks and hype in previous CPU releases. Something smells a little fishy to me......

I dunno, I think their marketing is doing just fine... This is hot news and every geek on the planet is desperately clawing over every scrap that comes out about the i7.... The hype is all over the place, every tech site worh its salt is running benchmarks on the CPU... and its barely costing intel a dime.....

Lets be honest who would intel be hypeing this too? Joe public who has a computer, yeah its that beige box in the dinning room we use to surf nuddie pics, shop on ebay, send email and play the occasional game on... The corperate buyers who buy by the 1,000 and just want a machine that will run MS Office as cheaply as possible... Or would it be aimed at us geeks who are getting all steamy under the collar and having to go for cold showers after we read the latest article telling us about the next great thing?

I think its called viral marketing by those people who Bill Hicks spoke so eloquently about, its just their new toy in the internet age... they don't need to spend millions, they just need to let the net and our own demand to have that perfect dream machine do the rest.

Intel aren't going to have problems selling the i7 when it hits the shelves.... they are going to have a problem getting enough chips on the shelves. By the time production of these bad boys is ramped up enough to meet demand the conroe will be selling for pennies most of the stock will be gone, or only produced because its cheap and still a good cpu.

[baius]

When I mentioned the hexus article I meant this page:
http://www.hexus.net/content/item.php?item=15015&page=8

I just believe the reason is not driver optimisation but the architecture, based on the musings from anand.

I guess we'll find out. (Link's dead, btw.)

I think its called viral marketing by those people who Bill Hicks spoke so eloquently about, its just their new toy in the internet age... they don't need to spend millions, they just need to let the net and our own demand to have that perfect dream machine do the rest.

Not all "leaks" are against company policy. There was a really good John Grisham book about that. Or was it Brad Meltzer?

Anyway, he portrayed the character of someone "high up" in Government having a "leak man" for news he wanted to get out, but not be traced back to him. (It was a good book...)